SkillsSkill 1 of 5
Read the cash flow behind the score
Measure coverage on the cash the business keeps, then size the loan to it before you look at collateral.
Worked example
The model scores this loan 82 out of 100. Would you approve it?
EBITDA coverage and collateral can make a loan look safe. Repayment comes from cash.
The question
Fernmoor Bank, a fictional regional bank, uses a model to score commercial loan requests. Calder Supply, an invented food-service distributor, asks for a $2.5m term loan to fund growth. Fernmoor’s policy minimum is 1.25 times debt service coverage, measured on cash available for debt service.
Risk score 82/100: low risk. Revenue up 20% to $12.0m. EBITDA covers pro forma debt service 1.60 times. Collateral covers the new loan 1.4 times. Recommendation: approve as requested.
| FY2024 | FY2025 | |
|---|---|---|
| Revenue | 10000 | 12000 |
| EBITDA | 1500 | 1800 |
| Cash taxes | 200 | 250 |
| Maintenance capex | 300 | 350 |
| Increase in working capital | 100 | 400 |
| Cash available for debt service | 900 | 800 |
| Existing debt service | 600 | 600 |
Invented data. Cash available for debt service is EBITDA less cash taxes, maintenance capex and the increase in working capital. The new loan, $2,500k at 7.0% repaid in equal annual payments over six years, adds $524k a year of principal and interest, so pro forma debt service is $1,124k.
- Relationship managerReceivables are up because two large customers moved to 75-day payment terms this year.
- OwnersWant the full $2.5m now. No new equity planned.
“The model scores Calder Supply 82 out of 100 and recommends approving a $2.5m term loan. Would you approve it as requested?”
Your turn
Say your answer out loud, or jot notes below. Both work. Or skip to the worked answer.
Not saved by this site. Your browser may restore text when you navigate back.
A strong answer, step by step
Open each step in turn. This is one strong answer, not the only one.
Step 1: Test the coverage the model used
What would you do here? Open the move you’d make.
Accept the model’s 1.60x: EBITDA of $1,800k comfortably covers $1,124k of debt service.
Tempting, but it misses something. EBITDA isn’t cash. After $250k of tax, $350k of capex and $400k into working capital, FY2025 left $800k, which covers $1,124k only 0.71 times.
Rebuild cash available for debt service: $800k after tax, capex and working capital covers $1,124k only 0.71 times, under the 1.25 minimum.
The strong move. The loan is repaid from cash, and on cash the request fails the policy that the 1.60 EBITDA figure seemed to pass.
Lead with the 1.4x collateral coverage: if the loan goes bad, the bank is covered.
Tempting, but it misses something. Collateral is the second way out. Leading with it skips whether $800k of cash can carry $1,124k of payments, which it can’t.
See how a strong candidate says it
“EBITDA of $1,800k covers $1,124k of debt service 1.60 times, but EBITDA isn’t cash. After tax, capex and working capital, FY2025 left $800k: 0.71 times the new debt service, well under our 1.25 times minimum.”
Why it works: The score summarized the income statement. The loan is repaid from cash.
Step 2: Ask whether the gap is temporary
What would you do here? Open the move you’d make.
Normalize working capital to FY2024’s $100k: cash of $1,100k still covers $1,124k only 0.98 times, so the request is too big.
The strong move. It separates a timing problem from a sizing problem. Even the generous case misses 1.25.
Treat the $400k working-capital build as one-off and use EBITDA less tax and capex, $1,200k.
Tempting, but it misses something. That assumes receivables never build again, and even $1,200k covers $1,124k just 1.07 times, still under 1.25.
Revenue grew 20% to $12.0m, so the cash squeeze is a sign of strength; approve to fund the growth.
Tempting, but it misses something. Growth is what used the cash: receivables rose as two large customers moved to 75-day terms. More debt doesn’t fix slower collections.
See how a strong candidate says it
“Working capital took $400k, up from $100k, because receivables grew. Even if it falls back to $100k, cash is $1,100k: 0.98 times. So slow collections don’t explain the whole shortfall. The request is too big.”
Why it works: It separates a timing problem from a sizing problem before deciding.
Step 3: Size what the cash can carry
What would you do here? Open the move you’d make.
Decline outright: coverage is below 1.25, so there’s nothing to lend.
Tempting, but it misses something. Existing debt alone is covered 1.33 times on FY2025 cash. The business can carry some new debt; the job is to size it.
Approve the full $2.5m because collateral covers the loan 1.4 times.
Tempting, but it misses something. Collateral protects the bank after a default. It doesn’t make $524k a year of new payments affordable.
Size to the cash: 1.25 times on $1,100k allows $880k of debt service; less $600k existing leaves $280k, about 53% of the $524k asked.
The strong move. It turns a no into a number the borrower can work with, and the collateral stays a second way out.
See how a strong candidate says it
“At 1.25 times on $1,100k, total debt service can be $880k. With $600k already committed, that leaves $280k a year for new debt, about 53% of the $524k requested. Collateral at 1.4 times is a second way out, not the first.”
Why it works: It turns a no into a number the borrower can work with.
Step 4: Structure it and watch the cause
What would you do here? Open the move you’d make.
Approve as asked with a 1.25x covenant tested on EBITDA, and review next year.
Tempting, but it misses something. Tested on EBITDA, the covenant starts at 1.60 and passes while cash covers 0.71. It would watch the wrong number.
Offer about half the term debt or a receivables-based line, with a 1.25x covenant on cash flow and a quarterly aging report.
The strong move. The structure follows the risk: the line grows with collected sales, and the aging report watches the two 75-day customers.
Approve the $2.5m but shorten the term so it’s repaid faster.
Tempting, but it misses something. A shorter term raises the annual payment above $524k and makes coverage worse, not better.
See how a strong candidate says it
“I’d offer about half the term debt, or a receivables-based line that grows with collected sales, with a 1.25 times coverage covenant tested on cash flow and a quarterly receivables aging report.”
Why it works: The structure follows the risk the numbers show, and the covenant watches it.
The answer, in short
- Decision
- Don’t approve $2.5m as requested. Offer about half the term debt, or a receivables-based line, with a cash-flow coverage covenant.
- Because
- On FY2025 cash the request covers debt service 0.71 times, and only 0.98 times with normal working capital. The model’s 1.60 times is EBITDA, not cash.
- What would change it
- Collections recovering so cash available for debt service clears 1.25 times the $1,124k of debt service, about $1.41m, or the owners putting in equity to shrink the loan.
Common traps
- Treating EBITDA as cash.
- Letting collateral approve a loan the cash can’t repay.
- Reading revenue growth as strength when it used up cash.
- Saying no without a structure that would work.
Check your answer
- Did I rebuild cash available for debt service from the table?
- Did I test whether the working-capital drain was temporary?
- Did I size the debt the cash can carry, not just reject it?
- Did I treat collateral as a secondary source of repayment?
Want feedback on how you’d answer? Practice “Read the cash flow behind the score” with an AI interviewer on Socratify.
Last reviewed Nov 9, 2026